Jump to content

Gerowen

Member
  • Posts

    83
  • Joined

  • Last visited

Reputation Activity

  1. Agree
    Gerowen got a reaction from leclod in Got Some New Seagate Exos Drives, Not Sure About Them...   
    Just filed for a return through Amazon and boxed the Seagates back up.
     
    Edit: When transferring a test file between the internal boot SSD and the RAID array where the network wouldn't be the bottleneck, I was getting read/write speeds in the several hundreds of MB/s, instead of 40.
  2. Agree
    Gerowen got a reaction from leclod in Got Some New Seagate Exos Drives, Not Sure About Them...   
    They were listed as "new" on Amazon.
     
    As far as the smart tests I ran short and conveyance tests both. After the resilver is finished in a couple hours I plan on running long ones.
     
    I installed Seagate's own "Seatools" and connected one of the drives and Seatools reported the same crazy values, but when I ran smartctl with a flag to convert them from 48 bit hex, the values were normal.
     
    I'm very seriously considering just returning all 3 and putting my old drives back in, ordering one more of them to upgrade with and calling it a day. Those WD drives report proper values in smart, they report their helium levels so you know if something has happened, they populate much faster so the whole system boots quicker and their read speeds don't tank during a resilver. Even on startup, 2 of the 3 Seagate drives take an unusually long time to spin up and start communicating. Whether it's my server or my laptop, when I connect one it's a good 30 seconds or more of the light flashing before the drives actually pop up and are ready to be accessed. This means that the server sits in the BIOS that whole time waiting on those drives to become responsive during POST.
  3. Like
    Gerowen got a reaction from TubsAlwaysWins in Post Linus Memes Here! << -Original thread has returned   
    So I just came across this on YouTube.  It "kinda" sounds like him, but it's not as convincing as that other channel called "There I Ruined It".  Anyway, just wanted to share.
     
     
  4. Like
    Gerowen reacted to AbydosOne in Considering the Switch to ZFS, Educate Me Please   
    No, it's still a pretty good rule regardless. Dedupe uses more than that, I'm pretty sure. That said, you're right that it's not mandatory.
     
    I don't think so, but I assume it's DDR2, which might begin to impede ZFS performance if you were really hammering it. My first FreeNAS server was a C2D with DDR2, and it was fine for storage purposes.
     
    Essentially zero. Yes, but leave it on. If a file is poorly compressible, it will abort and store it without compression.
     
    TBH, for what you're doing, I'd jump to TrueNAS Core or Scale (probably Scale as you seem familiar with Linux). You can run Plex and Nextcloud as plugins right out of the box. They (Scale and Core) both use ZFS natively and are tightly integrated with it.
     
    Just curious why you are bothering with encrypting drives on your own server?
     
    FWIW, I'm not really an expert on ZFS, but I've done my fair share of research and implementation on it, so feel free to ask follow-up questions.
  5. Like
    Gerowen reacted to Electronics Wizardy in Considering the Switch to ZFS, Educate Me Please   
    You don't need anywhere near that much ram, and tha rule is BS. I'd be fine using that array on 16GB of ram, or way less. ZFS is just using ram as a read cache, just like it does with ext4.
     
    It will probably be faster than compression. Even old CPUs can compress data pretty fast, and you can set the level of compression. I don't generally ses a reason to turn it off with slow hdds.
     
    No encryption and it still has issues with parity raid. If you can use ZFS I would here.
     
     
  6. Like
    Gerowen reacted to Electronics Wizardy in Considering the Switch to ZFS, Educate Me Please   
    Arguably the checksums don't help that much in ZFS as all your drives has checksums built into them, but if the drives or the interface is acting weird it can help in some cases where parity woudln't, like:
     
    -I think many raid cards only read the data and ignore parity. They would have no idea if the data was bad, checksums would fix this, and ZFS would try to use the parity to fix this.
    -IIf there are 2 copies of the data, and they disagree, there would be no way normally to know what copy is correct, checksums fix this.
  7. Informative
    Gerowen got a reaction from Lurick in My Steam Deck Killed my Internet...   
    I've had my Steam Deck and dock for probably a year or more now and this is the first, and only time this has ever happened, and usually yes, I shut it down or put it to sleep or something before disconnecting the dock.  I'll probably make sure to do that in the future just to be safe.
  8. Like
    Gerowen reacted to BlueChinchillaEatingDorito in My Steam Deck Killed my Internet...   
    Reminds me of the one time I got called into the animal shelter I volunteer for because their entire network went down. A quick Wireshark capture showed the credit card machine went absolutely nuts and flooded the network with DHCP requests the moment it was plugged in.
  9. Like
    Gerowen reacted to Alex Atkin UK in My Steam Deck Killed my Internet...   
    I have seen something similar which appeared to be a broadcast storm (every port on the switches were flashing like crazy), from my Sony surround receiver.  I started connecting it over WiFi instead of wired, haven't tried it again since.

    Its odd unplugging a USB ethernet adapter would cause it though as I would expect that to stop responding on the ethernet port once disconnected from the Steam Deck, though presumably its powered from a USB hub in the dock so was able to get stuck in a bad state.  I wonder if making sure you put the Deck to sleep first will help, or would that be worse?
  10. Informative
    Gerowen got a reaction from Angusticlavii in I Didn't Realize Just How Bad It Was... (FX 8370 CPU)   
    I've been kind of attached to the old girl just because it's worked reasonably well for so long. I tried to go beyond just locking it at its 4.3Ggz turbo and I've had it as high as 4.7Ghz, but even if it seems stable, it always eventually crashes. I'm debating on whether I lost the silicon lottery, or after watching LTT's recent video on power supplies, I've started wondering if my Thermaltake SMART 650w power supply is letting the voltage drop on the CPU power line when I try to push the overclock.
     
    My current home server/NAS is running an old AMD Phenom II x6 1045t, so what I may do is just migrate the FX chip into being my home server system for hosting Plex, Nextcloud and Minecraft, maybe give it a better power supply from Corsair or something and leverage that GPU for Plex encoding if I can. That Phenom does ok, but when there's 4 or 5 of us on the Minecraft server and somebody forces a chunk load you can tell it struggles a little for a few seconds and I feel like the FX would give us a bit more overhead in that kind of scenario. I could even leave the Blu-ray drive in it and just have the server do all of its own ripping and such.
  11. Informative
    Gerowen got a reaction from Angusticlavii in I Didn't Realize Just How Bad It Was... (FX 8370 CPU)   
    So about 7 years ago I set out to build myself the best AMD gaming rig I could with a budget of about a grand (ended up spending about $1,200).  This was a year-ish pre-Ryzen.  I ended up settling on the FX 8370 CPU, 16GB of RAM at 1866 Mhz (DDR3) and an MSI RX 480 8GB graphics card.  It did surprisingly well and I've enjoyed many great games on it.  Even today, I can fire up just about anything I want at 1080p with high settings and expect to get 60ish fps, or more, depending on the game.  There were some exceptions either way; DOOM (2016) gave me 100+ pretty much constantly at 1080p max settings, but Fallout 4 never wants to hit 60fps in down-town even when I turn down the graphics settings.  I knew that it wasn't the best CPU available even at the time, but it was 8 cores at a high clock speed for a reasonable price, and I wanted to do an all AMD build.  It was the first gaming PC I'd had in a very long time.  My last serious one before that had an NVidia GeForce FX 5500 GPU in it and I had been a console gamer since then, but when Sony ditched backwards compatibility with the PS4 and wouldn't even bring over all the digital stuff I bought on the PS Store through my PS3, they pushed me back to PC gaming.
     
    So over the years, I never noticed any major issues when it came to gaming on the rig.  Battlefield 1 would hitch occasionally, but for the most part, it stayed at or above 60fps and everything was great.  And some of that hitching stopped when I moved from Windows to Debian Linux so that Windows Update or the disk indexer weren't trying to fire up randomly on their own while I was gaming.  Anyway, fast forward a few years and I grabbed myself a Steam Deck.  I hadn't gamed seriously in some time because of work and life, and I thought that having the option to pick it up and take it with me would be nice.  Plus, I'm a Linux enthusiast, and I thought it was kind of awesome to see a main stream device of this nature from a well known company shipping with a Linux distro.  Since getting it, about 99% of my gaming has been done on the Deck.  Despite the lower GPU grunt compared to my desktop, it's nice to be able to sit in the bedroom or wherever and play my games, and even when I dock it, with some decent AA and sitting 8 or so feet from a 55 inch screen, I don't mind the  720p so much, and some games, like Halo MCC, can even hit 1080p with 100+ fps on the little thing.  Plus, the thing draws like 20-25 watts under load while docked, while my desktop will draw over 400 watts to play the same game.  But I did make one observation.  While the resolution was lower, the framerate seemed more consistent.  Battlefield 1 for example, hitched even less when there was a lot of explosions and such happening, it seemed more stable on the Deck.  So that got me to wondering, was the CPU in my Deck holding up better than the 125 watt FX-8370 in my desktop?
     
    So, I downloaded GeekBench on all of my devices, made sure no extraneous services like Steam were running in the background or anything, made sure everything's fans were working and that the CPUs were idling, and let them rip.  The results were startling to say the least.  I expected the Steam Deck to win in terms of CPU power, I had pretty much surmised as much based on my gaming experience.  What I didn't expect however, is that my Pixel 6a PHONE also thoroughly trounced the old FX-8370.  And this is with a minor overclock that put my FX-8370 markedly higher than the average score for an FX-9590, a 220 watt TDP chip that shipped with a water cooler.  So my cell phone, running on a couple of watts of battery power, the thing I use mostly just to send text messages and take photos, has more CPU grunt than a 125 watt desktop CPU from just a few years ago.  I knew the poor thing was struggling, but man, that's bad.  Even my Bluray rips for my Plex server have slowed to a crawl.  One of the desktop's only remaining tasks since getting the Deck has been to rip and transcode Bluray discs for storage on my Plex server, and I recently switched from x264 to x265, and the encode speed dropped tremendously to an average of 15ish fps in Handbrake.  These benchmark results make it pretty plain that the only reason it has done as well as it has at gaming is because of that RX-480 GPU.
     
    I think it's time to retire the old thing but honestly, since I'm doing all my gaming on the Deck, I'll probably just build something in a smaller case with an APU of some kind.  All I really need it for is that Bluray drive, so I don't need a big chonker sitting behind the TV with a big graphics card in it.  All I need is something that can turn a Bluray disc into an ISO and then convert that into a play-able file with Handbrake in a reason-able amount of time, and not suck down several hundred watts doing it.
     
    So anyway, here are the results from Geekbench.  The FX 8370 was the slowest thing I tested.
     
    FX-8370 OC'd to 4.3Ghz - Air cooled with stock AMD "Wraith" air cooler
    Single Core: 595
    Multi-Core: 2549
    https://browser.geekbench.com/v6/cpu/1945133
     
    Lenovo IdeaPad 330S w/ Ryzen 5 2500U - Hooked up to laptop chill pad for extra airflow
    Single Core: 1094
    Multi-Core: 3020
    https://browser.geekbench.com/v6/cpu/1945328
     
    Google Pixel 6a w/CalyxOS - Case removed and sitting over an air vent to try and make sure it didn't thermal throttle or anything
    Single Core: 1460
    Multi-Core: 3490
    https://browser.geekbench.com/v6/cpu/1945378
     
    Steam Deck - Docked with official dock
    Single Core: 1253
    Multi-Core: 4445
    https://browser.geekbench.com/v6/cpu/1945217
  12. Like
    Gerowen reacted to BiotechBen in PCIe USB Expansion Card Breaks Ethernet (Linux)   
    Have you considered going into BIOS to see if there are PCIE configuration settings such as 16x4, 8x8x4, and 4x4x4x4? Sounds like lane allocation is not happening correctly.
  13. Like
    Gerowen got a reaction from BiotechBen in My Steam Deck is On Its Way   
    So I got my email a couple days ago and my Steam Deck is on its way.  I'm super excited and just wanted to post this for two reasons mainly.
     
    A) To let you know it seems like a lot of people are getting their decks lately.  Gardiner Bryant (The Linux Gamer on YouTube) got the one he pre-ordered, and I personally know of two other people who got their emails a week or two before mine arrived, so it seems like Valve is catching up with the pre-orders.
     
    B) To ask if anybody knows if it comes with a screen protector.  I know the 512GB version comes with the etched glass screen and I'd lose out on that by putting a glossy screen protector on it, but I value protecting the screen more than the matte finish, but I don't want to order a screen protector if it already has one with it.
     

  14. Like
    Gerowen got a reaction from Needfuldoer in I have trouble screwing in my pc fans 120mm   
    +1 on this.  If the screws are screwing into the fans themselves, a lot of the fans don't have pre-cut threads and the screws literally cut threads into the plastic as they go, so the first time you install a particular fan, you've gotta use a little bit of force to get the screws tight.
  15. Agree
    Gerowen reacted to Needfuldoer in I have trouble screwing in my pc fans 120mm   
    Most of the fan screws I've encountered just cut their own threads into the holes in the fans. They take a worrying amount of effort compared to screws that meet machined threads in other parts, but that's normal.
     
    Try tightening them 1/4 turn, then back off 1/8 of a turn, then forward another 1/4, then back another 1/8. That should help keep the burr from binding the fresh threads up. Just don't overtighten, or you'll ruin the threads you just cut in the plastic and you'll have to come up with an alternate solution, like longer machine screws and nuts from the hardware store that can reach all the way through the case and the fan.
  16. Like
    Gerowen reacted to jagdtigger in The Russians Are Getting More Active...   
    @Gerowen
    There is nice plugin in nextcloud called GeoBlocker, set it to white list and only allow you country. (I dont see any of this since pfblocker-ng blocks russia, china, and asia [plus recently added ukraine])
  17. Informative
    Gerowen reacted to Levent in The Russians Are Getting More Active...   
    You can block entire IP ranges by few clicks if you are running a firewall. If not, you can always block manually. I personally have blocked all Russian, Chinese and Syrian IP ranges on all my network entry points.
    https://www.ip2location.com/free/visitor-blocker
  18. Like
    Gerowen reacted to duncannah in GoodWill ransomware forces victims to “transform themselves into a kind human being” to get their files back   
    I suppose if you're poor yourself, you're screwed? 
  19. Informative
    Gerowen got a reaction from AnnPc in Are USB/Flash Drives more reliable than HDD's?   
    I'm not sure about the longevity, but I know all the ones I have are SLOW, even when plugged into USB 3.1 ports.  Like 30-50 MB/s tops depending on the drive and the types of files being transferred.  My spinning rust drives hit around 110 MB/s.
     
    Another down-side is the flash modules in USB drives.  Yes, they don't have moving parts, but they would probably suffer from the same degradation as a regular SSD if you wrote to it often, probably worse since, as Schnoz says, they usually use inferior parts compared to regular SSDs and such.
     
    For cold storage though that isn't accessed often, I guess they'd be alright.
     
    If you want something in the 1-2 TB range that doesn't have moving parts, just get a normal SSD and stick it in an external enclosure.  Higher quality flash, higher speeds, better reliability.
  20. Informative
    Gerowen got a reaction from AnnPc in Are USB/Flash Drives more reliable than HDD's?   
    I have also noticed they're made of different materials, which might be part of the reason they're less reliable.  I've shot a handful of old drives before tossing them in the trash and with the 3.5" drives, the platters will just bend and warp around the bullet hole, but with 2.5" drives the platters shatter like glass and can fly everywhere.
  21. Agree
    Gerowen reacted to Electronics Wizardy in Accessing a shared network drive outside my home network   
    Setup a vpn? Use something like SCP to copy files over the wan?
     
    Technically you can port forward it, but you really don't want to port forward smb.
  22. Agree
    Gerowen reacted to Electronics Wizardy in Server with remote access VMs   
    How much performance do you need from the vms? Getting good performance will likely be a big pain here.
     
    Unraid really isn't made for the use you listed use as it has its way of doing raid thats isn't like traditional raid. Id probably go proxmox if you want a more tradition hypervisor.
     
    VMware, Citrix, and Microsoft all have a pretty good VDI solution that would work here. Problem is none are free and the cost in often in the thousands for a basic setup. So your kinda forced to use proxmox, setup the vms manually, then connect to them using parsec or simmilar.
     
    From my use the performance is pretty bad. Try watching a youtube video, I doubt it would playback smoothly. Also it doesn't use the gpu power of an attached gpu as its basically connected to the virtual gpu that the host created.
  23. Agree
    Gerowen got a reaction from da na in Are USB/Flash Drives more reliable than HDD's?   
    I'm not sure about the longevity, but I know all the ones I have are SLOW, even when plugged into USB 3.1 ports.  Like 30-50 MB/s tops depending on the drive and the types of files being transferred.  My spinning rust drives hit around 110 MB/s.
     
    Another down-side is the flash modules in USB drives.  Yes, they don't have moving parts, but they would probably suffer from the same degradation as a regular SSD if you wrote to it often, probably worse since, as Schnoz says, they usually use inferior parts compared to regular SSDs and such.
     
    For cold storage though that isn't accessed often, I guess they'd be alright.
     
    If you want something in the 1-2 TB range that doesn't have moving parts, just get a normal SSD and stick it in an external enclosure.  Higher quality flash, higher speeds, better reliability.
  24. Agree
    Gerowen got a reaction from Alex Atkin UK in Are USB/Flash Drives more reliable than HDD's?   
    I'm not sure about the longevity, but I know all the ones I have are SLOW, even when plugged into USB 3.1 ports.  Like 30-50 MB/s tops depending on the drive and the types of files being transferred.  My spinning rust drives hit around 110 MB/s.
     
    Another down-side is the flash modules in USB drives.  Yes, they don't have moving parts, but they would probably suffer from the same degradation as a regular SSD if you wrote to it often, probably worse since, as Schnoz says, they usually use inferior parts compared to regular SSDs and such.
     
    For cold storage though that isn't accessed often, I guess they'd be alright.
     
    If you want something in the 1-2 TB range that doesn't have moving parts, just get a normal SSD and stick it in an external enclosure.  Higher quality flash, higher speeds, better reliability.
  25. Agree
    Gerowen got a reaction from Poinkachu in Are USB/Flash Drives more reliable than HDD's?   
    I'm not sure about the longevity, but I know all the ones I have are SLOW, even when plugged into USB 3.1 ports.  Like 30-50 MB/s tops depending on the drive and the types of files being transferred.  My spinning rust drives hit around 110 MB/s.
     
    Another down-side is the flash modules in USB drives.  Yes, they don't have moving parts, but they would probably suffer from the same degradation as a regular SSD if you wrote to it often, probably worse since, as Schnoz says, they usually use inferior parts compared to regular SSDs and such.
     
    For cold storage though that isn't accessed often, I guess they'd be alright.
     
    If you want something in the 1-2 TB range that doesn't have moving parts, just get a normal SSD and stick it in an external enclosure.  Higher quality flash, higher speeds, better reliability.
×